daily-meeting-update
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources that could contain malicious instructions designed to influence the agent's behavior.\n
- Ingestion points: The skill pulls PR titles, commit messages, and review summaries via the GitHub CLI, ticket summaries via the Jira CLI, and session titles from local Claude Code history files in ~/.claude/projects.\n
- Boundary markers: The instructions do not specify the use of delimiters or protective wrappers when displaying or processing these external strings to distinguish them from trusted instructions.\n
- Capability inventory: The skill has the capability to execute shell commands (CLIs and a bundled Python script) and read local session history data.\n
- Sanitization: There is no evidence of explicit sanitization or filtering of the content retrieved from external integrations before it is presented to the agent.
Audit Metadata